US7203349B2 - Bronchial wall thickening recognition for reduced false-positives in pulmonary nodule detection - Google Patents

Bronchial wall thickening recognition for reduced false-positives in pulmonary nodule detection Download PDF

Info

Publication number
US7203349B2
US7203349B2 US10/059,848 US5984802A US7203349B2 US 7203349 B2 US7203349 B2 US 7203349B2 US 5984802 A US5984802 A US 5984802A US 7203349 B2 US7203349 B2 US 7203349B2
Authority
US
United States
Prior art keywords
airway
candidate
testing
pixels
airways
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related, expires
Application number
US10/059,848
Other languages
English (en)
Other versions
US20030144598A1 (en
Inventor
Li Zhang
Li Fan
Jianzhong Qian
Guo-Qing Wei
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Siemens AG
Original Assignee
Siemens Corporate Research Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Siemens Corporate Research Inc filed Critical Siemens Corporate Research Inc
Priority to US10/059,848 priority Critical patent/US7203349B2/en
Assigned to SIEMENS CORPORATE RESEARCH, INC. reassignment SIEMENS CORPORATE RESEARCH, INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: FAN, LI, QIAN, JIANZHONG, WEI, GUO-QING, ZHANG, LI
Priority to JP2003564800A priority patent/JP2005515874A/ja
Priority to AU2003205257A priority patent/AU2003205257A1/en
Priority to DE10392245T priority patent/DE10392245T5/de
Priority to PCT/US2003/001762 priority patent/WO2003065280A2/en
Priority to CN03802811.5A priority patent/CN1623160A/zh
Publication of US20030144598A1 publication Critical patent/US20030144598A1/en
Publication of US7203349B2 publication Critical patent/US7203349B2/en
Application granted granted Critical
Assigned to SIEMENS CORPORATION reassignment SIEMENS CORPORATION MERGER (SEE DOCUMENT FOR DETAILS). Assignors: SIEMENS CORPORATE RESEARCH, INC.
Assigned to SIEMENS AKTIENGESELLSCHAFT reassignment SIEMENS AKTIENGESELLSCHAFT ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: SIEMENS CORPORATION
Expired - Fee Related legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30061—Lung
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S128/00—Surgery
    • Y10S128/92—Computer assisted medical diagnostics
    • Y10S128/922—Computer assisted medical diagnostics including image analysis

Definitions

  • Pulmonary or lung cancer is currently a leading cause of cancer death. Early detection of cancer-related pulmonary nodules may provide the greatest chance to prevent deaths due to lung cancer.
  • Non-invasive, high-resolution, thin-slice, multi-slice or multi-detector computed tomography (“CT”) scanners are capable of providing vast quantities of detailed imaging data on anatomical structures. Therefore, non-invasive early detection of pulmonary nodules from CT images holds great promise.
  • a system and method for automatically reducing false-positive nodule candidates associated with airways that includes receiving a nodule candidate, testing for airway cavities connected to the candidate, and recognizing the candidate as a false-positive nodule candidate if it is connected to an airway cavity; where the testing may include perpendicular testing for airways that are relatively perpendicular to an examination plane and parallel testing for airways that are relatively parallel to an examination plane.
  • the present disclosure teaches an approach to reducing false-positives for pulmonary nodule detection.
  • FIG. 1 shows a block diagram of a system for automatically recognizing bronchial wall thickening from CT images according to an illustrative embodiment of the present disclosure
  • FIGS. 2A and 2B show CT image diagrams with false-positive nodules resulting from bronchial wall thickening for two typical airways;
  • FIG. 3 shows a schematic diagram for two extreme cases of airway appearance on scan planes
  • FIG. 4 shows a plot of airway branch orientation relative to a scan plane
  • FIG. 5 shows a flow diagram for automatically recognizing false-positive nodules from CT images according to an illustrative embodiment of the present disclosure
  • FIG. 6 shows a schematic diagram of anatomic airway features
  • FIG. 7 shows a flow diagram for performing a perpendicular test in an airway direction
  • FIG. 8 shows enlarged original images and thresholded images illustrating compactness and area testing
  • FIG. 9 shows a flow diagram for using adjacent vessel information in combination with an indeterminate perpendicular test
  • FIG. 10 shows a schematic diagram illustrating profile extraction and connection of high-confidence pixels in the parallel test
  • FIG. 11 shows a flow diagram for determining the overall confidence for a parallel airway
  • FIG. 12 shows a flow diagram for performing a parallel test to detect airways
  • FIG. 13 shows a flow diagram for detecting airways on examination planes.
  • CT computed tomography
  • a bronchial wall thickening detection technique incorporated into existing automatic nodule detection systems causes the false-positive ratio to be substantially reduced.
  • the present disclosure teaches an automatic bronchial wall thickening detection module that can be used with current nodule detection systems as a false-positive filter to identify the nodule candidates caused by bronchial wall thickening.
  • FIG. 1 shows a block diagram of a system 100 for automatically detecting pulmonary nodules from CT images, according to an illustrative embodiment of the present disclosure.
  • the system 100 includes at least one processor or central processing unit (“CPU”) 102 in signal communication with a system bus 104 .
  • CPU central processing unit
  • a read only memory (“ROM”) 106 , a random access memory (“RAM”) 108 , a display adapter 110 , an I/O adapter 112 , and a user interface adapter 114 are also in signal communication with the system bus 104 .
  • a display unit 116 is in signal communication with the system bus 104 via the display adapter 110 .
  • a disk storage unit 118 such as, for example, a magnetic or optical disk storage unit, is in signal communication with the system bus 104 via the I/O adapter 112 .
  • a mouse 120 , a keyboard 122 , and an eye tracking device 124 are also in signal communication with the system bus 104 via the user interface adapter 114 . The mouse 120 , keyboard 122 , and eye-tracking device 124 are used to aid in the detection of suspicious regions in a digital medical image.
  • a perpendicular testing unit 170 and a parallel testing unit 180 are included in the system 100 and in signal communication with the CPU 102 and the system bus 104 . While the perpendicular testing unit 170 and the parallel testing unit 180 are illustrated as coupled to the at least one processor or CPU 102 , these components are preferably embodied in computer program code stored in at least one of the memories 106 , 108 and 118 , wherein the computer program code is executed by the CPU 102 .
  • the system 100 may also include a digitizer 126 in signal communication with the system bus 104 via a user interface adapter 114 for digitizing a CT image of the lungs.
  • the digitizer 126 may be omitted, in which case a digital CT image may be input to the system 100 from a network via a communications adapter 128 in signal communication with the system bus 104 , or via other suitable means as understood by those skilled in the art.
  • FIGS. 2A and 2B CT images are shown with false-positive nodules resulting from bronchial wall thickening for two typical airways.
  • false-positive results are often caused by bronchial wall thickening due to the partial volume effect.
  • this disclosure teaches detection of airway cavities connected to nodule candidates.
  • the method uses two different approaches to identify two types of airway branches, which are based on geometric and gray level feature analysis of airways appearing in CT images.
  • FIG. 2A a pair of closely associated airway and vessel 210 are seen to be connected to a false-positive nodule 212 , where the pair 210 is perpendicular to an examination or scan plane (here, plane of page).
  • an examination or scan plane here, plane of page
  • a false-positive nodule 214 is shown where its airway 216 is parallel to the scan plane.
  • the features are identified by means of a perpendicular test and/or a parallel test.
  • the tube-like airway branches 310 and 316 are shown on slices in CT images with different appearances depending on the cutting angle between the scan planes and the airway branches.
  • Airway “B” 310 is perpendicular to the scan plane
  • Airway “A” 316 is parallel to the scan plane.
  • Most airway branches appear on scan planes between those two extreme cases.
  • the airway branches are divided into two types. First, if the axis direction of an airway branch is close to the perpendicular direction of the scan plane, the appearance on the scan plane is a disk or an ellipse surrounded by the airway wall. Second, if the axis direction of an airway branch is close to the parallel direction of the scan planes, the appearance on the scan plane is a strip with two nearly parallel airway walls.
  • FIGS. 2A and 2B give the two types of airways from real CT images.
  • the airway branch in pair 210 is perpendicular or nearly perpendicular to the scan plane; while in FIG. 2B , the airway branch 216 is parallel or nearly parallel to the scan plane.
  • FIG. 4 two different approaches are used to detect two types of airway branches because of the different geometric features of each.
  • the first approach called perpendicular testing, detects airway branches that are perpendicular or nearly perpendicular to the scan plane 400 using features derived from perpendicular airways.
  • the second approach is called parallel testing and recognizes parallel or nearly parallel airways.
  • FIG. 4 illustrates the cutting angle ranges 410 for the perpendicular test and 416 for the parallel test.
  • Function block 510 receives a nodule candidate, such as, for example, a nodule candidate resulting from the method of co-pending Ser. No. 10/008,133 entitled “Vessel-Feeding Pulmonary Nodule Candidate Generation”.
  • Decision block 512 executes a perpendicular test to determine whether the candidate is merely part of a relatively perpendicular airway wall. If the test result is true, the candidate is classified at function block 514 as a false-positive nodule due to bronchial-wall thickening.
  • decision block 516 executes a parallel test to determine whether the candidate is merely part of a relatively parallel airway wall. If this test result is true, the candidate is marked at function block 514 as a false-positive nodule due to bronchial-wall thickening. If the result of decision block 516 is negative, the candidate is classified at function block 518 as a potentially true nodule. Thus, if either the perpendicular test or the parallel test leads to a decision that this nodule candidate is located on an airway, the candidate is considered as a false-positive.
  • FIG. 6 anatomical features of airways are used for testing.
  • the perpendicular testing searches relatively perpendicular airways in a Region of Interest (“ROI”) by using the following anatomic knowledge: 1) Airway 618 lumens are dark regions in CT images; 2) Airway lumens are surrounded by relatively bright airway walls 620 in CT images; 3) Airways 618 and vessels 616 are often closely associated and branch in parallel while vessels 616 are much brighter in comparison with airway 618 lumen and lung parenchyma 622 .
  • FIG. 2A shows an exemplary pair 210 having a closely associated airway and vessel in a real CT image.
  • Nodule candidates are received at function block 710 .
  • the ROI is defined at function block 712 for the airway detection procedure based on the position of a nodule candidate being examined.
  • the ROI is thresholded with a predefined threshold, and a connected component analysis is used to obtain airway candidates.
  • Decision block 716 compares the compactness and area of these candidates with heuristic values for airways. If the test is false, function block 718 classifies the candidate as not located on an airway, typically by returning a logical false.
  • function block 720 performs an airway wall existence test to classify the probability of the candidate being part of an airway wall. If the probability is “weak”, control is passed to the above-described block 718 . If the probability is “strong”, control is passed to function block 722 to classify the candidate as located on an airway, typically by returning a logical true. However, if the probability is “middle”, an additional adjacent vessel test is performed at decision block 724 . If block 724 detects an adjacent vessel, the candidate is classified as located on an airway by the above-described block 722 . If block 724 does not detect an adjacent vessel, the candidate is classified as not located on an airway by the above-described block 718 . Thus, the distance to the potential airway from the nearest vessel are combined together to make a final decision about whether the nodule candidate is located on an airway relatively perpendicular to the scan plane.
  • a predefined global threshold is first selected to segment out possible airway voxels.
  • the Hounsfield number for air is ⁇ 1000 HU.
  • the threshold for airway lumens is approximately ⁇ 874 HU in this exemplary embodiment, although other comparable values may be used in alternate embodiments.
  • a connected component analysis is used to obtain regions that consist of mostly air, which become airway region candidates.
  • FIG. 8 two geometric features, namely compactness and area, are used in the perpendicular test to check airway region candidates.
  • an image 810 is thresholded using T airway of ⁇ 874 HU to produce a thresholded image 812 .
  • a nodule candidate 814 is found in both images. If an airway branch is perpendicular or nearly perpendicular to a scan plane, its appearance on the scan plane should be a disk or ellipse. Both disks and ellipses are highly compacted shapes. If compactness is defined as (perimeter) 2 /area, disks and ellipses should have relatively small compactness numbers.
  • a dark region has a large compactness number, it is very likely that this region is an arbitrary-shaped lung parenchyma region, or a relatively parallel airway region.
  • the area of airway regions is limited in a certain range to exclude random noises with small areas and lung parenchyma regions with large areas.
  • the airway region 816 is highly compact with a satisfactory area value. However, a further test is required in order to conclude that the airway region 816 is truly an airway region.
  • Another image 818 is thresholded to produce the thresholded image 820 .
  • the images 818 and 820 include a different nodule candidate 822 , and no connected airway regions are present. Accordingly, the nodule candidate 822 is a true-positive nodule candidate.
  • the confidence level for wall existence is labeled according to one of three descriptions: 1) If the number of bright pixels over the total number of outer boundary pixels is greater than a certain predefined value, for example 80%, the wall existence is “strong”; 2) If the number of dim pixels over the total number is greater than a certain predefined value, for example 50%, the wall existence is “weak”; 3) If the wall existence cannot be labeled with strong or weak, the wall existence is “middle”.
  • the nodule candidate 814 has a connected airway region 816 with a strong confidence level. Therefore, the candidate 814 is a false-positive nodule candidate.
  • the system detects vessels in the ROI and calculates the distance between the vessels and the airway candidate to assist decision making for airway existence.
  • the Hounsfield number for vessels in CT images can be approximated by the Hounsfield number for water, which is 0 HU.
  • vessel regions should be bright relative to lung parenchyma.
  • the area of vessel regions is preferably in a range chosen to exclude random noise and chest wall regions.
  • the adjacent vessel existence test is used.
  • the gray level of airway wall in CT images can vary in a large range. Therefore, to achieve noise insensitive and robust results, a decision is made by combining the confidence numbers of airway wall existence and of adjacent vessel existence in this case.
  • a gray-level confidence function 912 based on the average gray level of boundary pixels, is scaled by weighting factor alpha at multiplier 914 .
  • G 1 and G 2 are gray level thresholds for airway wall detection.
  • G 1 and G 2 are approximately 224 HU and 424 HU, respectively, in this exemplary embodiment, although other comparable values may be used in alternate embodiments.
  • a distance confidence function 916 based on the distance between the nearest vessel and the airway candidate, is scaled by weighting factor beta at multiplier 918 .
  • D 1 and D 2 are distance thresholds for the adjacent vessel test.
  • D 1 and D 2 are approximately 2.0 mm and 4.5 mm, respectively, in this exemplary embodiment, although other comparable values may be used in alternate embodiments.
  • a summing junction 920 receives the weighted confidence levels from the multipliers 914 and 918 , and provides the raw level to an overall confidence function 922 .
  • a decision block 924 receives the overall confidence level and determines whether it is greater than or equal to a threshold value T conf , which is approximately 0.75 in this exemplary embodiment, although other comparable values may be used in alternate embodiments. If the overall confidence is not less than T conf , function block 926 determines that the candidate is located on an airway. If the overall confidence is less than T conf , function block 928 determines that the candidate is not located on an airway.
  • the profile extraction and connection of high-confidence pixels in the parallel test is indicated generally by the diagram 1010 .
  • a ridge detector is used to calculate the perpendicular direction to the airway wall 1016 of the parallel airway 1014 with the nodule candidate 1012 .
  • profiles parallel to the direction from the ridge detector are extracted.
  • the profiles are analyzed to assign the parallel wall existence confidence to the middle pixel of the airway-like segments on the profiles.
  • a parallel airway wall confidence calculator 1110 has a first gray level confidence function 1112 for determining the confidence in a first airway wall section and weighting the result by a factor at a multiplier 1114 .
  • This weighting factor is approximately 0.5 in this exemplary embodiment, although other comparable values may be used in alternate embodiments.
  • a second gray level confidence function 1116 determines the confidence in a second airway wall section, and this result is weighted by a factor at a multiplier 1118 .
  • This weighting factor is approximately 0.5 in this exemplary embodiment, although other comparable values may be used in alternate embodiments.
  • a summing function 1120 receives the products of multipliers 1114 and 1118 , and supplies the sum to an overall confidence function 1122 for determining the overall confidence for the pair of airway walls.
  • a flow diagram 1210 of the parallel test for airway detection includes a function block 1212 for receiving preferably automatically detected nodule candidates, which feeds a function block 1214 .
  • the function block 1214 calculates a profile direction with a ridge detector, and feeds a function block 1216 .
  • the function block 1216 analyzes profiles and assigns confidence numbers for the middle pixel of each potential airway profile.
  • a function block 1218 follows block 1216 , connects pixels of high confidence number within a range of tolerance, and feeds decision block 1220 .
  • Decision block 1220 compares the number of connected pixels with a threshold N conn . If the number of connected pixels is less than N conn , function block 1222 determines that the pixel is not located on an airway.
  • decision block 1224 compares the average confidence of the connected pixels with a threshold T conf . If the average confidence is less than T conf , function block 1222 determines that the pixel is not located on an airway. If the average confidence is not less than T conf , function block 1226 determines that the pixel is located on an airway.
  • the parallel wall existence confidence is calculated from the gray level values of the both sides of a dark piece on the profiles. After confidence number calculation, pixels with high confidence numbers are connected into a contiguous line or curve with a certain tolerance, and a decision is made that the nodule candidate is located on a parallel airway.
  • airway tests may be performed on different viewing planes.
  • the flow diagram 1310 detects airways for nodule candidates on different spin planes, sagittal planes, and coronal planes.
  • a function block 1312 receives nodule candidates, preferably from an automatic detector, and passes control to a decision block 1314 .
  • the decision block 1314 detects airways on all spin planes by perpendicular and parallel testing, and passes control to function block 1316 if a corresponding airway is detected.
  • Function block 1316 determines that the nodule candidate is a false nodule caused by bronchial wall thickening.
  • decision block 1314 If decision block 1314 does not detect an airway, it passes control to decision block 1318 , which detects airways on sagittal planes using both perpendicular and parallel testing. If block 1318 detects an airway, it passes control to function block 1316 , as above. However, if it does not detect an airway, it passes control to decision block 1320 to detect airways on coronal planes. If decision block 1320 detects an airway, it passes control to function block 1316 , as above. If decision block 1320 does not detect an airway either, it passes control to a function block 1322 , which determines that the nodule candidate is considered to be a true nodule candidate upon bronchial wall thickening detection.
  • the nodule candidate is considered as a false-positive caused by bronchial wall thickening, phlegm, or dirt accumulated at an airway bifurcation point.
  • the final outputs of this system can be either directly provided to physicians on the display devices by visually marking the false-positive lung nodules, by providing the list of detected false-positives to a CAD system to automatically remove the false-positives and to improve the overall diagnostic accuracy of such CAD systems, or by providing an updated list of true candidates to a nodule detection system, such as, for example, one described in co-pending Ser. No. 10/008,119 Vessel-Feeding Pulmonary Nodule Detection By Volume Projection Analysis”.
  • the present disclosure teaches automatically filtering false-positive nodule candidates resulting from bronchial wall thickening and related phenomenon from CT images so that radiologists and physicians can be freed from the heavy burden of reading through multitudes false-positive nodule candidates.
  • An advantage of the present disclosure is the provided sensitivity to pulmonary nodules while maintaining low false-positive rates.
  • pulmonary nodules appear in slice images as nearly circular-shaped opacities, which are similar to cross-sections of vessels. Accordingly, many existing recognition methods have a high false-positive rate.
  • the present disclosure solves this problem by detecting airways and their associated false-positive nodule candidates. It shall be understood that, although exemplary embodiments have been described with reference to CT imaging, the present disclosure is also applicable to other types of imaging, such as, for example, to magnetic resonance imaging (“MRI”).
  • MRI magnetic resonance imaging
  • teachings of the present disclosure may be implemented in various forms of hardware, software, firmware, special purpose processors, or combinations thereof. Most preferably, the teachings of the present disclosure are implemented as a combination of hardware and software.
  • the software is preferably implemented as an application program tangibly embodied on a program storage unit.
  • the application program may be uploaded to, and executed by, a machine comprising any suitable architecture.
  • the machine is implemented on a computer platform having hardware such as one or more central processing units (“CPU”), a random access memory (“RAM”), and input/output (“I/O”) interfaces.
  • CPU central processing units
  • RAM random access memory
  • I/O input/output
  • the computer platform may also include an operating system and microinstruction code.
  • various processes and functions described herein may be either part of the microinstruction code or part of the application program, or any combination thereof, which is executed via the operating system.
  • various other peripheral units may be connected to the computer platform such as an additional data storage unit and a printing unit.

Landscapes

  • Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Image Processing (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)
US10/059,848 2002-01-29 2002-01-29 Bronchial wall thickening recognition for reduced false-positives in pulmonary nodule detection Expired - Fee Related US7203349B2 (en)

Priority Applications (6)

Application Number Priority Date Filing Date Title
US10/059,848 US7203349B2 (en) 2002-01-29 2002-01-29 Bronchial wall thickening recognition for reduced false-positives in pulmonary nodule detection
PCT/US2003/001762 WO2003065280A2 (en) 2002-01-29 2003-01-21 Bronchial wall thickening recognition for reduced false-positives in pulmonary nodule detection
AU2003205257A AU2003205257A1 (en) 2002-01-29 2003-01-21 Bronchial wall thickening recognition for reduced false-positives in pulmonary nodule detection
DE10392245T DE10392245T5 (de) 2002-01-29 2003-01-21 Erkennung von Bronchialwanderverdickungen zur Verringerung falsch positiver Befunde beim Nachweis von Lungenknoten
JP2003564800A JP2005515874A (ja) 2002-01-29 2003-01-21 肺の小結節検出時のフォールスポジティブを低減するための気管支壁厚認識
CN03802811.5A CN1623160A (zh) 2002-01-29 2003-01-21 降低肺结节检测中假阳性率的支气管壁增厚识别

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US10/059,848 US7203349B2 (en) 2002-01-29 2002-01-29 Bronchial wall thickening recognition for reduced false-positives in pulmonary nodule detection

Publications (2)

Publication Number Publication Date
US20030144598A1 US20030144598A1 (en) 2003-07-31
US7203349B2 true US7203349B2 (en) 2007-04-10

Family

ID=27609907

Family Applications (1)

Application Number Title Priority Date Filing Date
US10/059,848 Expired - Fee Related US7203349B2 (en) 2002-01-29 2002-01-29 Bronchial wall thickening recognition for reduced false-positives in pulmonary nodule detection

Country Status (6)

Country Link
US (1) US7203349B2 (de)
JP (1) JP2005515874A (de)
CN (1) CN1623160A (de)
AU (1) AU2003205257A1 (de)
DE (1) DE10392245T5 (de)
WO (1) WO2003065280A2 (de)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050063579A1 (en) * 2003-09-18 2005-03-24 Lee Jeong Won Method of automatically detecting pulmonary nodules from multi-slice computed tomographic images and recording medium in which the method is recorded
US8675934B2 (en) 2010-04-30 2014-03-18 Vucomp, Inc. Breast skin line detection in radiographic images
US8675933B2 (en) 2010-04-30 2014-03-18 Vucomp, Inc. Breast segmentation in radiographic images
US9256799B2 (en) 2010-07-07 2016-02-09 Vucomp, Inc. Marking system for computer-aided detection of breast abnormalities
US9265458B2 (en) 2012-12-04 2016-02-23 Sync-Think, Inc. Application of smooth pursuit cognitive testing paradigms to clinical drug development
US9380976B2 (en) 2013-03-11 2016-07-05 Sync-Think, Inc. Optical neuroinformatics

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7116810B2 (en) * 2002-11-27 2006-10-03 General Electric Company Method and system for airway measurement
WO2004077358A1 (en) * 2003-02-28 2004-09-10 Cedara Software Corporation Image region segmentation system and method
US8442280B2 (en) * 2004-01-21 2013-05-14 Edda Technology, Inc. Method and system for intelligent qualitative and quantitative analysis of digital radiography softcopy reading
CN101443813B (zh) * 2006-05-17 2012-07-04 皇家飞利浦电子股份有限公司 输出造影增强的医学图像数据的计算机程序
EP2122571B1 (de) * 2006-12-19 2018-02-21 Koninklijke Philips N.V. Vorrichtung und verfahren zur anzeige von wahrscheinlichen computerdetektierten falschen positivergebnissen in medizinischen abbildungsdaten
EP2252214A4 (de) * 2008-02-13 2013-05-29 Kitware Inc Verfahren und system zur messung von gewebeschäden und krankheitsrisiko
US9349176B2 (en) * 2009-07-15 2016-05-24 Mayo Foundation For Medical Education And Research Computer-aided detection (CAD) of intracranial aneurysms
WO2015183934A1 (en) * 2014-05-29 2015-12-03 Ohio State Innovation Foundation Volumetric texture score

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6141437A (en) * 1995-11-22 2000-10-31 Arch Development Corporation CAD method, computer and storage medium for automated detection of lung nodules in digital chest images
US20010044576A1 (en) * 1994-10-27 2001-11-22 Vining David J. Method and system for producing interactive three-dimensional renderings of selected body organs having hollow lumens to enable simulated movement through the lumen
US20020164060A1 (en) * 2001-05-04 2002-11-07 Paik David S. Method for characterizing shapes in medical images
US20020193687A1 (en) * 1994-10-27 2002-12-19 Vining David J. Automatic analysis in virtual endoscopy
US6549646B1 (en) * 2000-02-15 2003-04-15 Deus Technologies, Llc Divide-and-conquer method and system for the detection of lung nodule in radiological images
US20030095696A1 (en) * 2001-09-14 2003-05-22 Reeves Anthony P. System, method and apparatus for small pulmonary nodule computer aided diagnosis from computed tomography scans
US20030105395A1 (en) * 2001-12-05 2003-06-05 Li Fan Vessel-feeding pulmonary nodule candidate generation

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20010044576A1 (en) * 1994-10-27 2001-11-22 Vining David J. Method and system for producing interactive three-dimensional renderings of selected body organs having hollow lumens to enable simulated movement through the lumen
US20020193687A1 (en) * 1994-10-27 2002-12-19 Vining David J. Automatic analysis in virtual endoscopy
US6141437A (en) * 1995-11-22 2000-10-31 Arch Development Corporation CAD method, computer and storage medium for automated detection of lung nodules in digital chest images
US6549646B1 (en) * 2000-02-15 2003-04-15 Deus Technologies, Llc Divide-and-conquer method and system for the detection of lung nodule in radiological images
US20020164060A1 (en) * 2001-05-04 2002-11-07 Paik David S. Method for characterizing shapes in medical images
US20030095696A1 (en) * 2001-09-14 2003-05-22 Reeves Anthony P. System, method and apparatus for small pulmonary nodule computer aided diagnosis from computed tomography scans
US20030105395A1 (en) * 2001-12-05 2003-06-05 Li Fan Vessel-feeding pulmonary nodule candidate generation

Non-Patent Citations (8)

* Cited by examiner, † Cited by third party
Title
Chabat F. et al., << ERS Transform for the Automated Detection of Bronchial Abnormalities on CT of the Lungs, >> IEEE Trnasactions on Medical Imaging, IEEE Inc. vol. 20 No. 9 2001, pp. 942-952 XP001100471.
Fan L. et al. "Automatic Detection of Lung Nodules From Multi-Slice Low-Dose CT Images," Proceedings of the Spie, vol. 4322, pp. 1828-1835 XP008022813.
Hoffman A. Eric, "Physiology and Function from Multidimensional Images," Proceedings SPIE-The International Society for Optical Engineering,vol. 2168 1994.
International Search Report.
Maryellen L. Giger, Kyongtae T. Bae, Heber Macmahon, "Computerized Detection of Pulmonary Nodules in Computed Tomography Images," Investigative Radiology, vol. 29, No. 4 1994, pp. 459-465, XP008024742.
Milan Sonka, Gopal Sundaramoorthy, Eric A. Hoffman, "Knowledge-Besed Segmentation of Intrathoracic Airways from multidimensional High Resolution CT Images," Proceedings SPIE, Medical Imaging 1994, Physiology and Function From Multidimensional Images, 1994, pp. 73-85, XP008024795.
P. Armstrong, A. G. Wilson, P. Dee, D.N. Hansell, "Imaging of Diseases of the Chest" Mosby, XP002266478 3<SUP>rd </SUP>edition, chapter 4, pp. 152-153.
Thinh V. Nguyen, Jack Sklansky, "Computing the Skeleton of Coronary Arteries in Cineangiograms," computers and Biomedical Research, Academic Press, Inc. vol. 19, 1986, pp. 428-444 XP008024914.

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050063579A1 (en) * 2003-09-18 2005-03-24 Lee Jeong Won Method of automatically detecting pulmonary nodules from multi-slice computed tomographic images and recording medium in which the method is recorded
US8675934B2 (en) 2010-04-30 2014-03-18 Vucomp, Inc. Breast skin line detection in radiographic images
US8675933B2 (en) 2010-04-30 2014-03-18 Vucomp, Inc. Breast segmentation in radiographic images
US8855388B2 (en) 2010-04-30 2014-10-07 Vucomp, Inc. Microcalcification detection classification in radiographic images
US8923594B2 (en) 2010-04-30 2014-12-30 Vucomp, Inc. Spiculated malignant mass detection and classification in radiographic image
US8958625B1 (en) 2010-04-30 2015-02-17 Vucomp, Inc. Spiculated malignant mass detection and classification in a radiographic image
US9076197B2 (en) 2010-04-30 2015-07-07 Vucomp, Inc. Probability density function estimation
US9256941B2 (en) 2010-04-30 2016-02-09 Vucomp, Inc. Microcalcification detection and classification in radiographic images
US9262822B2 (en) 2010-04-30 2016-02-16 Vucomp, Inc. Malignant mass detection and classification in radiographic images
US9256799B2 (en) 2010-07-07 2016-02-09 Vucomp, Inc. Marking system for computer-aided detection of breast abnormalities
US9265458B2 (en) 2012-12-04 2016-02-23 Sync-Think, Inc. Application of smooth pursuit cognitive testing paradigms to clinical drug development
US9380976B2 (en) 2013-03-11 2016-07-05 Sync-Think, Inc. Optical neuroinformatics

Also Published As

Publication number Publication date
WO2003065280A2 (en) 2003-08-07
WO2003065280A3 (en) 2004-03-25
AU2003205257A1 (en) 2003-09-02
CN1623160A (zh) 2005-06-01
JP2005515874A (ja) 2005-06-02
US20030144598A1 (en) 2003-07-31
DE10392245T5 (de) 2005-03-10

Similar Documents

Publication Publication Date Title
US8050481B2 (en) Method and apparatus for small pulmonary nodule computer aided diagnosis from computed tomography scans
CN101231678B (zh) 医用图像处理装置及医用图像处理方法
US6154560A (en) System and method for staging regional lymph nodes using quantitative analysis of endoscopic ultrasound images
US7203349B2 (en) Bronchial wall thickening recognition for reduced false-positives in pulmonary nodule detection
US20020028008A1 (en) Automatic detection of lung nodules from high resolution CT images
El-Baz et al. Three-dimensional shape analysis using spherical harmonics for early assessment of detected lung nodules
JP5784404B2 (ja) 画像処理装置、画像処理方法、及び画像処理プログラム
US20050207630A1 (en) Lung nodule detection and classification
US20080002870A1 (en) Automatic detection and monitoring of nodules and shaped targets in image data
Zong et al. Automatic ultrasound image segmentation based on local entropy and active contour model
CN112052896B (zh) 图像处理方法及装置，分类模型的训练方法及装置
CN112365973B (zh) 基于对抗网络和Faster R-CNN的肺结节辅助诊断系统
US20100111392A1 (en) System and method for automatically classifying regions-of-interest
US7711172B2 (en) Automatic nodule measurements
US7020316B2 (en) Vessel-feeding pulmonary nodule detection by volume projection analysis
Kanazawa et al. Computer aided screening system for lung cancer based on helical CT images
CN113139928A (zh) 肺结节检测模型的训练方法和肺结节检测方法
US8160336B2 (en) Reducing false positives for automatic computerized detection of objects
US20050002548A1 (en) Automatic detection of growing nodules
US20030105395A1 (en) Vessel-feeding pulmonary nodule candidate generation
US7369638B2 (en) System and method for detecting a protrusion in a medical image
JP2012504003A (ja) コンピュータを用いて実行される障害検出方法及び装置
JPH08320923A (ja) 異常陰影自動検出システム
Mohamed et al. Computer aided diagnosis of digital mammograms
Horsthemke et al. Predicting LIDC diagnostic characteristics by combining spatial and diagnostic opinions

Legal Events

Date Code Title Description
AS Assignment

Owner name: SIEMENS CORPORATE RESEARCH, INC., NEW JERSEY

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ZHANG, LI;FAN, LI;QIAN, JIANZHONG;AND OTHERS;REEL/FRAME:012708/0705

Effective date: 20020307

AS Assignment

Owner name: SIEMENS CORPORATION,NEW JERSEY

Free format text: MERGER;ASSIGNOR:SIEMENS CORPORATE RESEARCH, INC.;REEL/FRAME:024185/0042

Effective date: 20090902

FPAY Fee payment

Year of fee payment: 4

AS Assignment

Owner name: SIEMENS AKTIENGESELLSCHAFT, GERMANY

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:SIEMENS CORPORATION;REEL/FRAME:031099/0819

Effective date: 20130805

REMI Maintenance fee reminder mailed
LAPS Lapse for failure to pay maintenance fees
STCH Information on status: patent discontinuation

Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362

FP Lapsed due to failure to pay maintenance fee

Effective date: 20150410